Sir,

Extrapolating from our findings that mRNA for the dermcidin gene (Schittek et al, 2001), which encodes the proteolysis-inducing factor (PIF) core peptide (PIF-CP), is present in both gastrooesophageal tumour and adjacent non-tumour tissue, EA Rapoport suggests that PIF may be generally involved in muscle breakdown via production of PIF in tissues requiring amino acids for tissue regeneration, for example, in liver regeneration. There are, however, several caveats, which need to be applied to this suggestion. The first is that it is only the glycosylated form of the PIF-CP that has been shown to induce muscle breakdown (Todorov et al, 1996), and this form is difficult to identify in tissues (Wieland et al, 2007). Glycosylated PIF has also been shown to have pro-inflammatory effects on hepatocytes (Watchorn et al, 2001) and other cell types. In our study (Deans et al, 2006), we made no claim that glycosylated PIF was present in gastrooesophageal tumours or adjacent tissues. In addition, the PIF-CP is homologous to a neural survival peptide (YP-30) (Cunningham et al, 2002), and expression of the gene encoding the PIF-CP has been shown to provide a small survival and proliferative advantage to other cell types including tumours (Lowrie et al, 2006, and Stewart et al, 2007). Furthermore, evidence suggests that the YP-30/PIF-CP peptide may have a role in development, acting as a maternal blood-borne factor, which promotes survival of the developing thalamus (Landgraf et al, 2005). In a recent development, there is now evidence that peptide products of the dermcidin gene may participate in the regulation of placental function by means of modulating proteolytic cascades on the trophoblastic surface (Motoyama et al, 2007) and having proteolytic activity. Therefore, while EA Rapoport's suggestion may be unlikely in the light of present evidence, there is still much to be discovered about this fascinating gene and its products.